The following abbreviations are herewith defined, at least some of which are referred to within the following description of the state-of-the-art and the present invention.
BLERBlock Error RateCPICHCommon Pilot ChannelCQIChannel Quality InformationCRCCyclic Redundancy CheckGSMGlobal System for Mobile CommunicationsHS-DSCHHigh-Speed Downlink Shared ChannelHSDPAHigh-Speed Downlink Packet AccessISInterim StandardLTELong Term EvolutionMCSModulation and Coding SchemeMIMutual InformationSINRSignal to Interference Plus Noise RatioSNRSignal to Noise RatioTDRFTransmit Diversity with Rich FeedbackWCDMAWideband Code Division Multiple Access
In the wireless communications field, there are a number of advanced antenna techniques/configurations that can be considered for current and future base station deployments, ranging from calibrated diversity antennas that can be used for coherent combining, to antennas that can be co-phased for adaptive directional steering, to fixed beam configurations with either analog or digital beam-forming. One aspect present in current base stations is their ability to improve coverage for mobiles by using channel quality feedback from the mobiles when performing link adaptation and scheduling of the mobiles. While the type of channel quality feedback may be different for different antenna configurations, each antenna configuration must be complementary to the use of the channel quality feedback to improve coverage. However, each type of antenna configuration has disadvantages that need to be minimized without causing mismatches for either the link adaptation process or user scheduling.
For instance, base stations that deploy calibrated diversity antennas may have channel state information fed back from the mobile so the base station can calculate transmission weights for each transmit antenna such that the subsequent transmission is matched to the channel. This scenario is denoted as closed-loop transmit diversity or transmit diversity with rich feedback (TDRF), and is designed for use with transmit antennas that do not exhibit high correlation between elements (e.g. antennas with large spatial separation or different'polarizations). Consider, for example, an approach similar to the above but for the high-speed downlink packet-data access (HSDPA) mode of a WCDMA communication system. One problematical issue when using this particular approach is described next which occurs when there are many mobiles in the WCDMA communication system. The base station sets transmit weights according to a scheduled (i.e. desired) mobile's channel state information (e.g. an estimate of the channel) that was fed back to the base station. During the data transmission to the scheduled user, the unscheduled mobiles in turn compute their channel state feedback (e.g. an estimate of the channel) after the uplink feedback transmission. However, a difficulty that arises is the SINR which is measured by the unscheduled mobiles on the common pilot channel (CPICH) may be different than the SINR achieved on the high-speed downlink shared channel (HS-DSCH). This difficulty arises because the transmit weights may be different for each channel, causing a different interference footprint to be present, affecting the SINRs. As a result, there is a mismatch between the achievable and measured SINRs on the CPICH and HS-DSCH channels. Consequently, the base station would need to account for this when selecting the modulation and coding schemes (MCSs) for the unscheduled mobiles. This topic is described in co-assigned U.S. Patent Application Publication No. US 2005/0136840 A1 entitled “SIR Estimates for Non-Scheduled Mobile Terminals” published on Jun. 23, 2005 (the contents of which are hereby incorporated herein by reference).
Alternatively, base stations that deploy co-phased antennas with closely spaced array elements can be used to adaptively steer energy towards a desired mobile and to place nulls in other directions. This is similar to the above approach, only now the co-phasing of the array establishes the directionality of the incoming and outgoing signals. This traditional beam-forming approach can obtain direction information from the uplink for creating the steering weights for the downlink transmission. While channel state information need not be fed back from the mobile, some measure of the channel quality must be sent, such as SINR or MCS information. This is commonly called the channel quality information (CQI) and since it is based on the SINR it may be subject to similar mismatches as was described above due to the adaptive nature of the beam-forming. The aforementioned U.S. Patent Application Publication No. US 2005/0136840 A1 also applies to this topic as well since the main problem is one of the CQI changing unexpectedly.
In another alternative, base stations that deploy fixed beam antenna configurations may have a number of beams placed within a cell or sector with fixed steering directions. One of these beams is selected for transmission to a particular mobile based on some criterion that is measured by either the mobile or the base station. For instance, the mobile may measure the signal strength transmitted along pilot channels in each beam, and select the strongest beam which the base station should use for subsequent transmission. Alternatively, measurements made during uplink transmissions, such as the signal strength in the best uplink beam, or the direction of the mobile can be used to select a downlink beam for transmission. While simpler from an antenna system perspective, there are still a number of issues concerning fixed beams that relate to CQI reporting which are similar to those discussed above with respect to calibrated diversity antennas and the co-phased antennas. Plus, there are a number of issues concerning fixed beams that relate to coverage which exist with respect to the fixed nature of the beam pattern. These issues are described in more detail next with respect to FIGS. 1 and 2 (PRIOR ART).
Referring to FIG. 1 (PRIOR ART), there is a block diagram illustrating the basic components of a fixed-beam wireless communication system 100 with mobiles 102 (only three shown) and a traditional base station 104 which includes a fixed beam antenna network 106, a scheduler unit 108, and a user data queue 110 (note: one having ordinary skill in the art will readily appreciate that descriptions of well-known components have been omitted so as not to obscure the current discussion). Basically, the scheduler unit 108 receives user information 112 (from the user data queue 110 which itself receives user data 114) and CQI 116 (from the mobiles 102) and then processes this information 112 and 116 to choose a specific mobile 102a (for example) to transmit user data 114 and from which fixed beam to transmit the user data 114 to the specific mobile 102a. In particular, the scheduler unit 108 processes information 112 and 116 to choose the specific mobile 102a (user selection 118) and then determine the transmission rate 120 and a beam selection 122 which are used to transmit the user data 114 to the mobile 102a. As shown, the scheduler unit 108 forwards the user selection 118 and the transmission rate 120 to the user data queue 110 and forwards the beam selection 122 to the fixed beam antenna network 106. The fixed beam antenna network 106 creates the fixed beams using analog and/or digital beam-forming.
Referring to FIG. 2 (PRIOR ART), there is a plot showing an example of a fixed beam pattern emitted from the traditional base station 104 where the beam pattern has four orthogonal beams 200a, 200b, 200c and 200d spanning over 180 degrees, although the coverage area might actually include a subset of the 180 degrees (e.g. the center 120 degrees). The beam gain for this particular fixed-beam pattern is plotted for each of the four beams vs. the direction. The following characteristics can be observed from this ideal fixed beam pattern:                There is a loss of roughly two dB at the cross-over points between the two center beams.        The beams have an unequal main-beam width as a function of angular beam (i.e. mobile) direction.        The high side-lobe levels of one beam can cause interference to mobiles in other beams, and this is most pronounced at the beam cross-over points.        The element pattern is multiplicative with the array pattern and causes a loss in beam-forming gain as a function of angular beam direction.        
In the past, these issues have been addressed in different manners. For instance, more narrow beams can be added to reduce the cross-over loss and the beam-shape can be changed as a function of direction. Plus, non-orthogonal (i.e. tapered) beams can be added to improve coverage at the cross-over points and to improve the side-lobe levels. However, the approach results in a loss in peak gain. Another approach is to vary the beam-shape for a particular beam over time so that it has the same average gain for any mobile direction. This approach is considered and discussed in co-assigned Swedish Patent Application No. 010381703 filed Nov. 16, 2001 (the contents of which are hereby incorporated herein by reference).
However, the proposed solutions to these issues do not address the aforementioned CQI measurement problems when beam selection is used for improving downlink coverage. In particular, the problems associated with measuring CQI can cause beams to be pointed to a desired (i.e. scheduled) mobile but potentially causes interference to unscheduled mobiles. This is because unscheduled mobiles may be located within the main-beam direction or in the side-lobe direction, and the interference will be different in each case. Additionally, when CQI is based on SINR levels, then the CQI will also vary depending on the mobiles' locations with respect to the fixed beam centers. Thus, the goal of obtaining uniform coverage should also consider obtaining more accurate/predictable CQI values when addressing the aforementioned issues associated with the fixed beam antenna configuration. These problems and other problems are satisfied by the present invention.